I WOULD like to raise some questions about how we do science. I
don鈥檛 know the answers, but what worries me is that when I discuss them with
colleagues I get the sense that no one does. It appears that in the past years
everyone but us academics has given a lot of thought to how their workplaces are
organised. So perhaps it would be worthwhile for scientists to start talking
about how we govern our communities and universities. The following is then
offered as a conversation starter.
Universities are fundamental to the project of democracy because they are the
main home of intellectuals who examine society and reflect on the possibilities
for change. As such, they are responsible for educating an increasing number of
our citizens, and, ideally, instructing them in the habits of reflection,
self-criticism and openness to others that are essential for participation in a
modern democracy.
It鈥檚 odd, then, that universities themselves are not terribly democratic in
their governance, and are quite slow to change. Indeed, quite a few academics
believe that the existing structure is the one to best fulfil the mission of the
university, and claim that to best serve the democracies that support them,
universities themselves should not be very democratic. I do not know whether
they are right or wrong, but here are some questions we could ask.
Advertisement
First of all, how did the present organisation of our universities arise? The
modern university evolved from schools of theology. These were designed to
perpetuate old knowledge and to promote the cultivation of wisdom on ancient
questions. Is this really the best setting for science, which aims to discover
new knowledge and to promote new questions?
Many of the institutions of modern society鈥攂usiness, industry,
government agencies鈥攈ave evolved dramatically in recent decades. Indeed,
there are whole professions devoted to debating, advising and managing change in
the business and government world. Even military men and women are nowadays
trained to consider issues of management, organisation and decision-making, as
well as the relationship between leaders and those they lead. Many of the people
who work on organisational change teach at universities, yet very little
discussion and reflection about structure and organisation has been aimed at the
universities themselves. Is this good or bad?
Businesses have been eliminating as many levels of management as they can,
opening up and flattening their organisations so they can capitalise quickly on
new ideas, technologies and products. Universities appear to have been going in
the opposite direction. Where once there was a department head or dean, now
there are often several suites of offices (usually nicer than those of the
faculty) filled with associate and assistant heads and deans, each of whom will
generally have secretaries and assistants.
One consequence has been that those who make decisions in universities spend
an increasing proportion of their time talking to other administrators rather
than to the teachers, researchers and students whom their decisions affect.
Worse, they became answerable to other administrators rather than to those
working to fulfil the mission of the organisation. These are exactly the trends
that modern business has sought to eliminate because they make an organisation
resistant to change. Shouldn鈥檛 the universities be doing the same?
Many scientists believe deeply in the hierarchical organisation of their
communities. There seems to be good reason to have a probationary period, after
which successful scientists gain the intellectual freedom that tenure gives. But
does this imply that scientists as a community cannot be democratic?
While there is obviously an important role for the kind of wisdom that
requires decades of experience to acquire, is concentrating power in the hands
of the more senior scientists best for science and scientists? Most big
decisions in the scientific community involve predicting the significance of
research a person has yet to produce. It is easy to recognise success, but
university positions, tenure and research grants are not prizes: they are what
enables people to do science. So the key question when hiring or funding a
scientist is what will he or she produce.
Most new ideas and new research programmes are initiated by younger
scientists, typically in the first decade of their career. But almost all the
power in the scientific community is in the hands of senior scientists in the
last decade or two of their career. Might this hold back scientific
progress?
One result of the present hierarchical structure is that it is mainly older
scientists who determine the distribution of resources, both between different
fields and between different approaches within a given field. As a result,
funding decisions often reflect academic politics rather than objective and
open-minded evaluations of scientific merit. Another consequence is that the
distribution of funding often fails to reflect the priorities of younger
scientists. Sometimes a very popular field will have much less funding than one
that leaves the best young scientists cold. This places economic pressure on
young scientists to go along with the research agendas of older generations of
scientists.
So is science hobbled by the fact that the scientists with the most energy,
enthusiasm, openness and flexibility鈥攚ho are therefore most in touch with
new ideas and developments鈥攈ave the least power when it comes to the
distribution of resources?
If the older scientists are in charge, it follows that science cannot
progress so quickly that those near the end of their careers do not understand
the work done by the postdocs. This suggests that theories and techniques
typically remain in use for around 40 to 50 years, and that only over longer
timescales does science change so radically that whole new languages and
techniques are invented. Does this imply that science might progress faster at
least some of the time if it were more democratic, so that scientists at all
stages of their careers had an equal say in decisions affecting appointments,
funding and distribution of resources?
Is there any evidence to support the assumption that scientists become better
at predicting the significance of their colleagues鈥� work as they get older? Yes,
experience clearly plays a part and many people do acquire wisdom with age. At
the same time, many a young academic will know the work of a job candidate
better than their older colleagues.
Any modern management consultant will tell you that people at different
levels of a hierarchy have very different views of how that organisation works,
and sometimes even on who is doing the key work on which everyone鈥檚 success
depends. They will say that the healthiest organisation is one which makes
decisions based on input from people at all levels. When I, as a professor, have
insisted that my department or research centre is not very hierarchical, the
consultants have countered that how hierarchical an organisation is and how open
those at the top are to the views of people below them are precisely the things
that are perceived differently up and down the hierarchy. How well do we
professors really do in seeking the views of our younger colleagues? How often
might we have made an ultimately better, if riskier, appointment if our younger
colleagues had participated as equals?
Of course, the mixture of hierarchy and democracy varies between countries.
In many universities in the US and Canada, department chairs are elected, and
hiring and promoting staff requires the vote of the whole department. In
Britain, heads are appointed and hiring is done by committees of external senior
scientists. In some institutes in Italy and France, the director of the
institute and all the directors of groups are elected by the staff. German
institutes are run by directors appointed by external committees who alone hold
the power for further appointments.
What is most striking is that the scientists in each country are often
convinced that their system is the only rational one, and that the quality of
the science they do is a consequence of the system they use. In Britain, for
example, there are sincere and thoughtful people who are convinced that it would
be a disaster to allow anyone to participate in the appointment of anyone of
equal or higher rank to them, while in Italy and France equally thoughtful
people believe the quality of the science done at their institutes would
decrease rapidly if directors were appointed rather than elected. But has anyone
done any research into the actual consequences of these different systems?
It must also be emphasised that science is not a pure meritocracy. The
quality, boldness and ambition of a young scientist鈥檚 work are not all that
determine their career success. Also important are the popularity of the area
they work in and their approach, the 鈥減edigree鈥� of their education and training,
and the influence and power of their supervisors and mentors. Good or bad, this
undoubtedly has some deleterious effects.
Chief among them is the cynicism generated when young scientists see that
quality and originality are not always rewarded as highly as conformity to the
research programmes of powerful elders. Science itself may lose out when young
people of unusual foresight and courage who choose to invent and work on their
own research are unable to obtain good jobs, while merely competent people in
popular areas鈥攖hose favoured by powerful, older scientists鈥攁dvance
easily.
If we really believe that science should be a meritocracy, and that
scientific progress depends on young people who think for themselves and
introduce new ideas rather than work on the ideas of their elders, shouldn鈥檛
they be the ones who are chosen for the really good jobs?
I once asked a venture capitalist how people in that business decided the
ideal level of risk to take on. He replied that experience told them that
long-term returns are maximised when 10 per cent of the new businesses they
support succeed. In my field, the proportion of papers that represent real
progress is certainly at best 10 per cent. But we write and evaluate grant
applications as if the good people can guarantee almost 100 per cent success.
This means that we successful scientists do achieve close to 100 per cent
success鈥攂ut for projects that are generally less ambitious and risky than
we dreamed about as graduate students.
While some of this represents the wisdom that science is a craft where
success is often incremental and little steps that succeed are as necessary as
bold ventures, we cannot delude ourselves that this is the whole story. More
often than we care to admit, many years of incremental steps lead either to a
dead end or to knowledge that turns out to be only incidental. If we could be
more honest about the risks with ourselves and those who back us financially,
could we make the science we attempt bolder and more ambitious? By taking more
risks in our own work, could we increase the overall rate of progress?
For the health of science, we clearly must encourage courageous scientists,
who invent and work on their own ideas rather than what is fashionable. And to
make true progress we must increase rather than limit the diversity of
approaches to the truly hard problems.
My last question is simple. Has anyone got any idea exactly how we do all
this?